The Federal Cartel Office (“FCO”) has published a background paper “Innovation – Challenges for the Antitrust Law Practice” in October 2017. It covers possible innovation aspects across different antitrust law areas, with a focus on merger control.
The paper is a valuable contribution to the debate on innovation. It is an agency paper, so it does not necessarily include criticism at established case-law. But it illustrates the uncertainties, challenges and open questions regarding cases concerning innovation unrelated to existing product markets, as was the case in Dow/DuPont, and new developments in the internet economy.
I. Innovation in economic theory
The first part deals with innovation from an economic perspective. It explores possible sources for innovation, innovation effects on market structure, and possible effects of market structure, of changes in market structure as well as of antitrust enforcement on innovation incentives. In particular, the paper looks at the results of economic research regarding possible effects of the intensity of competition on innovation incentives.
The paper concludes that innovation and market structure are interdependent, and that there is not much, and maybe not sufficient, economic literature available to properly capture the effects of this interaction. (This is mainly because the economic models used in existing literature are predominantly static instead of complex and dynamic, which the paper suggests would be more suitable). The paper is almost a call for further research in this area. One challenge is how to measure innovation in quantitative and qualitative aspects.
According to the paper many currently available research results seem to suggest that elimination of competition chills innovation incentives, at least in already highly concentrated markets. However, in certain circumstances increased market concentration may also have positive effects on innovation incentives. This can lead to tensions between static and dynamic analysis. For example, short term negative effects of a merger or a cooperation, like price increases, may have to be balanced with positive long-term effects like increased innovation incentives. Closely connected is the question how to weigh static vs. dynamic effects in antitrust cases. The paper concludes that there is no general answer.
II. Innovation in legal antitrust analysis in practice
The paper provides an overview of the relevant current antitrust practice of the European Commission and the FCO. Main parts include the analysis of (i) existing product markets; (ii) future product markets; and (iii) competition in innovation unrelated to specific products. The paper shows that the further the analysis drifts away from existing product or technology markets, the more difficult to establish a case. The analysis of competition in innovation unrelated to specific products is a new development and still largely unexplored territory. The paper concludes that generally there are still many open and unresolved question for discussion (and provides a catalogue of questions at the end).
1. Existing product markets
a) Reduced innovation as theory of harm
The paper describes well-established theories of harm in this area, including elimination of actual competition in innovation between the parties of a horizontal merger (GE/Alstom), or hampering rivals’ innovation in the downstream product market following a vertical merger (Intel/McAfee). The same can apply to cooperation agreements and be found in abuse of dominance cases, e.g., Microsoft (tying the Internet Explorer to Windows) re. foreclosing competitors and thus chilling innovation.
A theory of harm in the context of potential competition is that the acquirer would cease or delay development of its own competing product (or vice versa) post-merger, and the Commission has applied this theory in various pharma merger cases (e.g., in Pfizer/Hospira) when the potential competitor’s pipeline products were in a very late development stage.
The introduction of an additional merger threshold (transactional value of €400 million) in Germany follows a similar theory of harm, i.e., that an established acquirer aims at eliminating the innovation potential of the target in order to protect its own portfolio.
b) Innovation may reduce market power
The paper acknowledges that innovation in a dynamic market may also undermine market power. This may depend on the market phase – in growing markets innovation may lead to quick market entry and shifting of market shares. Other factors may be the duration of product life-cycles and technological development.
In the internet economy, market power based on indirect network effects (digital platforms) may be reduced by the innovative force of the internet. One important question is whether the disruptive, innovative nature of the internet may typically render market power in this area less stable. The paper concludes that there is no general answer, but the need for a case-by case analysis, including other possible factors resulting in disruptive change to undermine market power. In Microsoft/Skype innovation in communications technology and short product lifecycles spoke against the parties’ market power (despite a combined share of 90%). The FCO similarly cleared the merger of two leading online dating platforms, which continued to face strong innovation pressure post-merger.
The latest competition law reform in Germany has introduced the criterion “innovation-driven competitive pressure” into a catalogue for assessing market power in two-sided markets with network effects. However, the paper stresses the difficulty in practice to determine how likely resulting market share shifting is.
2. Competition in future product markets
Future product markets may be relevant when the development activities relate to products that would create new product markets. In various pharma merger cases the Commission took into account novel products in advanced R&D stage in order to properly assess the competitive situation, as did the FCO in Tokyo Electron/Applied Materials (regarding future production technology markets).
The paper concedes that it is a challenge to determine the likelihood of success of R&D activities and whether they would lead to product launches. In addition, it is more difficult to assess the competitive situation of future than of existing markets, given that the future market conditions, including suppliers, customers and potential entry barriers (e.g., patents following successful innovation), are unknown.
3. Innovation activities unconnected to specific product markets
The paper shows that this is a new development in the legal analysis in practice and the most challenging scenario. If R&D activities cannot be assigned to particular products (yet), the change of market structure may nevertheless have negative effects on innovation and on competition in potential future, not yet identified markets. The paper correctly raises the question how possible effects can be adequately predicted and considered in the antitrust analysis.
The paper refers to Dow/DuPont, in which the Commission analyzed effects on innovation itself, speaking of innovation competition and innovation spaces instead of product/price competition and product markets, respectively. The theory of harm was the delay, termination or redirection of the parties’ overlapping R&D lines and early pipeline products, as well as an overall structural reduction of incentives for R&D in the entire industry.
The paper concludes that taking into account R&D activities unrelated to specific products in the antitrust analysis raise several theoretical and practical questions: this approach does not distinguish between R&D efforts and innovation, even though more R&D does not necessarily mean more innovation, given that the R&D’s success depends on numerous other factors, like quality, focus, etc. One challenge is to determine the competitive position of companies in innovation – expenditure may be one aspect which, however, does not equal success. The Commission considered addition aspects, including the commercial success of the parties’ innovation or the number of their patents and patent quotations. The paper additionally refers to economic theory that shows that under certain circumstances increased concentration may also be beneficial for innovation incentives. If applicable, this would need to be balanced with possible negative effects.